University of Groningen

Molecular Characterization of Staphylococcus aureus Isolates Transmitted between Patientswith Buruli UlcerAmissah, Nana Ama; Chlebowicz, Monika A.; Ablordey, Anthony; Sabat, Artur J.; Tetteh,Caitlin S.; Prah, Isaac; van der Werf, Tjip S. ; Friedrich, Alexander; van Dijl, Jan; Rossen,JohannesPublished in:PLoS Neglected Tropical Diseases

DOI:10.1371/journal.pntd.0004049

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2015

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Amissah, N. A., Chlebowicz, M. A., Ablordey, A., Sabat, A. J., Tetteh, C. S., Prah, I., ... Stienstra, Y. (2015).Molecular Characterization of Staphylococcus aureus Isolates Transmitted between Patients with BuruliUlcer. PLoS Neglected Tropical Diseases, 9(9), [0004049]. DOI: 10.1371/journal.pntd.0004049

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 11-02-2018

RESEARCH ARTICLE

Molecular Characterization of Staphylococcusaureus Isolates Transmitted betweenPatients with Buruli UlcerNana Ama Amissah1,2*, Monika A. Chlebowicz3, Anthony Ablordey2, Artur J. Sabat3,Caitlin S. Tetteh2, Isaac Prah2, Tjip S. van der Werf1, Alex W. Friedrich3, Jan Maarten vanDijl3, JohnW. Rossen3☯, Ymkje Stienstra1☯

1 Department of Internal Medicine/Infectious Diseases, University of Groningen, University Medical CenterGroningen, Groningen, the Netherlands, 2 Department of Bacteriology, Noguchi Memorial Institute forMedical Research, University of Ghana, Legon, Ghana, 3 Department of Medical Microbiology, University ofGroningen, University Medical Center Groningen, Groningen, the Netherlands

☯ These authors contributed equally to this work.* namaamissah@noguchi.ug.edu.gh

Abstract

Background

Buruli ulcer (BU) is a skin infection caused byMycobacterium ulcerans. The wounds of

most BU patients are colonized with different microorganisms, including Staphylococcusaureus.

Methodology

This study investigated possible patient-to-patient transmission events of S. aureus duringwound care in a health care center. S. aureus isolates from different BU patients with over-

lapping visits to the clinic were whole-genome sequenced and analyzed by a gene-by-gene

approach using SeqSphere+ software. In addition, sequence data were screened for the

presence of genes that conferred antibiotic resistance.

Principal Findings

SeqSphere+ analysis of whole-genome sequence data confirmed transmission of methicil-

lin resistant S. aureus (MRSA) and methicillin susceptible S. aureus among patients that

took place during wound care. Interestingly, our sequence data show that the investigated

MRSA isolates carry a novel allele of the fexB gene conferring chloramphenicol resistance,

which had thus far not been observed in S. aureus.

Author Summary

Buruli ulcer (BU) is a skin infection caused byMycobacterium ulcerans. The wounds ofmost BU patients are colonized with different microorganisms, including Staphylococcus

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 1 / 12

OPEN ACCESS

Citation: Amissah NA, Chlebowicz MA, Ablordey A,Sabat AJ, Tetteh CS, Prah I, et al. (2015) MolecularCharacterization of Staphylococcus aureus IsolatesTransmitted between Patients with Buruli Ulcer. PLoSNegl Trop Dis 9(9): e0004049. doi:10.1371/journal.pntd.0004049

Editor: Pamela L. C. Small, University of Tennessee,UNITED STATES

Received: May 13, 2015

Accepted: August 11, 2015

Published: September 11, 2015

Copyright: © 2015 Amissah et al. This is an openaccess article distributed under the terms of theCreative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in anymedium, provided the original author and source arecredited.

Data Availability Statement: All relevant data arewithin the paper and its Supporting Information file.

Funding: NAA was supported by fellowships fromthe Graduate School for Medical Sciences of theUniversity of Groningen. YS was supported by aVENI grant from the Netherlands Organisation forScientific Research and the Gratama foundation. Thefunders had no role in study design, data collectionand analysis, decision to publish, or preparation ofthe manuscript.

aureus. This study investigated patient-to-patient transmission events during wound carein a health care center. S. aureus isolates from patients who visited the health center at thesame time points were analyzed using whole-genome sequencing. Analysis of sequencedata confirmed transmission of methicillin resistant S. aureus and methicillin susceptibleS. aureus among patients that took place during wound care.

IntroductionBuruli ulcer (BU) is a neglected necrotizing skin disease caused byMycobacterium ulcerans,with the highest burden of the disease in West Africa, particularly in Benin, Cote d’Ivoire andGhana [1]. The disease usually starts as a nodule, plaque, oedema or papule and progresses toform large ulcers with undermined edges if left untreated. It was previously shown that woundsof most BU patients are heavily colonized by many microorganisms, including Staphylococcusaureus [2,3].

S. aureus can be part of the human microbiota colonizing the skin and mucosal membraneswithout any clinical manifestations. However, once it crosses the skin barrier, or when the hostimmune system is compromised, this bacterium is able to cause a wide range of diseases, suchas skin and soft tissue infections, osteomyelitis, pneumonia, meningitis, or bacteremia [4,5].Therefore, S. aureus is considered a dangerous pathogen in both community-acquired and nos-ocomial infections. Colonization of healthy individuals with multi-drug resistant S. aureus isregarded as a risk factor for future development of S. aureus infections that are difficult to treat[6]. The S. aureus colonization of patients with a serious breach of skin barrier, such as patientswith BU, burn wounds or the group of hereditary mechanobullous diseases epidermolysis bul-losa (EB), was previously shown to be very high [2,3,7–9]. Molecular typing of S. aureus iso-lated from the wounds of BU and EB patients has shown that their wounds often harbormultiple genotypes of this pathogen [3,10].

Recently, several S. aureus clones have been reported in health care institutions in Ghanawith the sequence types (ST) 15, 121 and 152 being the most prevalent as determined by multi-locus sequence typing (MLST) [11]. Notably, health care-associated infections (HAIs) causedby S. aureus impose a significant burden on patient care as a result of prolonged hospital stays,increased cost of treatments and high morbidity and mortality rates. Current practices imple-mented to reduce HAIs include cleaning of the hospital environment, hand hygiene andscreening and decolonization of patients and health care workers [12–15]. Epidemiologicaldata and molecular typing methods, such as pulsed-field gel electrophoresis (PFGE), MLST,spa-typing, multiple-locus variable number tandem repeat fingerprinting (MLVF), and whole-genome sequencing (WGS) of the infecting strains can be used to trace transmission events[16–19]. Each of these typing methods has particular advantages [20]. For example, MLVF isfast, cheap and highly discriminatory [21], while WGS provides additional information on thegenetic makeup of investigated isolates on top of a highly discriminatory typing result.

BU patients may be at risk of hospital-associated colonization with S. aureus due to theirfrequent visits to particular health care centers for wound care. This represents an additionalhealth risk for these patients, even if they are already colonized with community-acquired S.aureus. Therefore, the present study was aimed at uncovering possible S. aureus transmissionevents among BU patients using MLVF andWGS. Furthermore, WGS was applied to identifyantimicrobial resistance (AMR) genes and to screen for mutations in genes that confer certainresistance phenotypes. The results obtained underpin the potential of the combined use ofMLVF and WGS for the surveillance of S. aureus outbreaks in hospital settings.

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 2 / 12

Competing Interests: The authors have declaredthat no competing interests exist.

Materials and Methods

Ethics statementThe ethical committee of the Noguchi Memorial Institute for Medical Research (NMIMR)(FEDERALWIDE ASSURANCE FWA 00001824) approved the use of clinical samples for thisinvestigation. Samples were collected upon written informed consent from adult subjects and aparent or guardian of any child participant on their behalf.

Bacterial isolates and genomic DNA extractionA subset of the S. aureus isolates from BU patients that were previously collected and groupedby MLVF into thirteen clusters (A-M) [3] were selected for WGS. For the present study, iso-lates were selected from each of the thirteen MLVF clusters including two clusters suspected ofpatient-to-patient transmission events during wound care (clusters H and F). Screening of BUpatients for the presence of S. aureus had been repeated every two weeks for a period of sevenmonths, which defined the sampling time points t1 to t13 in this study (Table 1). Patientsinvolved in this screening were at different stages of the disease and treatment for BU. All pres-ently investigated S. aureus isolates were obtained from positive anterior nares and wound cul-tures of eleven BU patients who attended the Pakro Health Center in the Eastern region ofGhana for antimicrobial therapy (Table 1).

Genomic DNA was extracted from S. aureus isolates grown overnight on blood agar byusing the Ultraclean microbial DNA isolation kit (mo bio laboratories, Inc, Carlsbad, Califor-nia, USA) according to the manufacturers’ instructions.

Table 1. Frequency of S. aureuswith different spa-types isolated during patient visits for wound care.

12-12-2012

9-1-2013

23-1-2013

6-2-2013

20-2-2013

6-3-2013

20-3-2013

3-4-2013

17-4-2013

9-5-2013

23-5-2013

20-6-2013

4-7-2013

PatientNo.

Start date oftreatment

t1 t2 t3 t4 t5 t6 t7 t8 t9 t10 t11 t12 t13

2 1-12-2012 p t786 t786 x p x x t786 t786 x x x x

ST88 7 26-12-2012 p p t355 t786 t084,t939

x t084,t939,t1096

x x t002,t084

x x

19 5-1-2013 p t786 x x x x x x x x x x

ST152 7 26-12-2012 p p t355 t786 t084,t939

x t084,t939,t1096

x x t002,t084

x x

5 22-12-2012 p p p p x x x x x t355 x x

6 12-12-2012 t7835 x x x x x x x x x t355 x x

3 6-12-2012 p t084 p p x x t084 t084,t355

t084,t314

t084 t084 p p

10 12-12-2012 p t355 t355 t355 t355 x x t355 x x x x x

11 19-12-2012 t355 x x t355 x x x x x x x x

18 7-1-2013 p p p t355 p p p x x x x x

24 14-2-2013 t355 x x x x x x x x

‘p’ indicates that a patient visited the health center, but no S. aureus was detected.

‘x’ indicates that a patient did not visit the health center at the respective time point of sampling.

‘t1’ to ‘t13’ refers to the time points at which samples were collected.

Cells with bold formatting indicate involvement of the respective S. aureus isolates in transmission events. spa-types with italic formatting are isolates that

were sequenced.

doi:10.1371/journal.pntd.0004049.t001

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 3 / 12

Whole-genome sequencing, sequence assembly and data analysesDNA libraries were prepared using the Nextera XT v2 kit (Illumina, San Diego, CA, USA)according to the manufacturers’ instructions and then run on a Miseq (Illumina) for generatingpaired-end 250-bp reads. De novo sequence assembly was performed using CLC GenomicsWorkbench v7.0.4 (CLC bio A/S, Aarhus, Denmark) after quality trimming (Qs> 28) withoptimal word sizes based on the maximum N50 value. The assembled files were imported asFasta files into SeqSphere+ software version 1.1 (Ridom GmbH). The sequence reads were sub-mitted to the National Center for Biotechnology Information GenBank and are available underthe BioProject PRJNA283747 and accession numbers: LGAE00000000, LFTW00000000,LFTV00000000, LFTU00000000, LFTT00000000, LFOH00000000, LFOG00000000,LFNS00000000, LFNR00000000, LFNQ00000000, LFNP00000000, LFNO00000000,LFNN00000000, LFNM00000000, LFNL00000000, LFNK00000000, LFNJ00000000,LFNI00000000, LFNH00000000, LFMH00000000, LFMG00000000. The sequence data of the21 isolates were characterized by using the core genome multilocus sequence typing (cgMLST)consisting of 1,861 genes and 706 S. aureus accessory genes. The complete sequence of eachisolate was analyzed based on gene-by-gene comparison with the reference S. aureus strainCOL (GenBank accession no. NC_002951) and S. aureus cgMLST target definer function withthe default parameters of the software as previously described [22]. Each allele was assigned anumber and an allelic typing profile based on the combination of all alleles for each isolate bythe software. A dendrogram of the sequenced isolates and two additional reference genomesthat represent different sequence types (ST5 [N315 GenBank accession no. BA000018.3] andST8 [COL]) was generated using an unweighted-pair group method using average linkages(UPGMA). The concordance between the two typing methods was calculated with the RidomEpiCompare software version 1.0 as described previously [19].

Transmission eventsIn this study a transmission event is defined to have occurred if the wound of a patient, previ-ously not containing a particular S. aureus genotype, becomes colonized over time by an S.aureus with a genotype that is identical with the genotype of an S. aureus isolate collected fromthe wound of another patient. Here we investigated whether transmission events had indeedoccurred during wound care of patients treated in the Pakro Health Center, using theSeqSphere+ scheme which assigned each S. aureus isolate an allelic typing profile as previouslydescribed. The typing profile will, subsequently, be known as cluster type (CT). Hence a trans-mission event would have occurred if S. aureus isolates from different BU patients are groupedwithin the same CT. In our previous study, S. aureus isolates from BU out-patients, who visitedthe health center for wound care, were suspected to be involved in patient-to-patient transmis-sion events [3]. These isolates were initially grouped by MLVF into clusters H and F (Fig 1A).

Screening for antimicrobial resistanceDe novo assembled genome sequences of S. aureus isolates were queried against specific previ-ously identified sequence features, or compared to complete S. aureus reference genomes withassociated annotated genes (S1 Table) using blastN in the WebACT comparison tool withdefault settings (http://www.webact.org/WebACT/prebuilt#). Further detailed analyses wereperformed with the Artemis Comparison Tool (ACT) software [23]. Specifically sequence datawere queried for the presence of SCCmec elements and AMR genes. Similarity matches werefiltered based on their length and percentage similarity scores, and only the filtered hits with atleast 80% sequence similarity were then displayed by ACT and analyzed in detail. The AMRgenes that were screened confer resistance to chloramphenicol, clindamycin, erythromycin,

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 4 / 12

Fig 1. MLVF and SeqSphere+ dendrograms of 21 S. aureus isolates from BU patients. The dendrograms were generated using the UPGMA algorithm.(A), Dendrogram showing the previously identified MLVF clusters A-M. (B), SeqSphere+ dendrogram showing the 14 CTs identified in the present study. The

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 5 / 12

fusudic acid, kanamycin, lincosamide, methicillin, mupirocin, penicillin, rifampicin, strepto-gramin A and B, tetracycline, trimethoprim, tobramycin, and/or vancomycin. Antibiotic resis-tance profiles of the sequenced isolates were previously determined using vitek according tothe EUCAST guidelines [3].

Results

Phylogeny of sequenced S. aureus isolates found in BU patients inGhana based on a gene-by-gene comparisonFrom a total of 13 BU patients who visited the Pakro Healthcare Center for wound care 21 S. aureusisolates from the anterior nares (n = 4) and wounds (n = 17) were sequenced. These included sixmethicillin resistant MRSA and 15 methicillin susceptible S. aureus (MSSA) isolates. These isolateshave been previously characterized byMLVF and spa-typing as shown in Table 2 [3].

A dendrogram was generated after SeqSphere+ analysis of the 21 sequenced isolates, whichrevealed 14 cluster types (CTs) (Figs 1B and 2) denoted as 545 (n = 4), 714 (n = 1), 556 (n = 1),546 (n = 1), 552 (n = 1), 554 (n = 1), 715 (n = 1), 549 (n = 2), 550 (n = 1), 553 (n = 1), 555(n = 1), 547 (n = 4), 551 (n = 1) and 548 (n = 1). This clustering by SeqSphere+ seemed tomatch well with the previous clustering of isolates by MLVF. To calculate the concordancebetween the SeqSphere+ and MLVF typing data, the Ridom epicompare software 1.0 (RidomGmbH) was used with the Rands Adjusted co-efficient. This revealed a concordance of 0.924.

The 21 sequenced isolates were assigned to eight MLST types, namely ST1, ST5, ST15,ST88, ST121, ST152, ST508, and the new ST3019. The ST3019 is a single-locus variant of ST45at the yqiL locus.

Evidence of patient transmission eventsA first transmission event was identified for four MRSA isolates belonging to ST88, which werepreviously grouped in the MLVF cluster H (Fig 1A) [3]. These isolates were classified bySeqSphere+ as CT 545 (Fig 1B). Within CT 545 the allelic profiles were identical (Fig 2). Ofnote, the four isolates were obtained from three different patients visiting the healthcare centerover a period of seven months. The medical care for these patients involved antibiotic treat-ment and wound dressing changes (Tables 1 and 2). This particular MRSA was first identifiedin the wound of patient 2, who tested negative at the first sampling time point (t1). Patient 2was the first to start treatment in this study and was found to carry this particular S. aureusgenotype at several sampling time points during treatment (i.e. at t2, t3, t8 and t9). Patients 7and 19 started treatment 25 and 35 days later, respectively. They both visited the health carecenter for wound care at time point t2. The wounds of patients 7 and 19 tested positive for S.aureus with the genotype of CT 545 at the sampling time points t5 (patient 7) and t3 (patient19; Table 1), which is indicative of transmission events.

A second suspected transmission event was initially identified by MLVF typing (cluster F)and involved eight BU patients [3]. To investigate this possible transmission event in moredetail four of the 25 isolates obtained from three patients were randomly selected andsequenced. These four MSSA ST152 isolates were assigned to CT 547 (Fig 1B). The allelic pro-files within this cluster differed by one (Fig 2). Patients 10 and 11 tested positive for S. aureuswith this particular genotype at the same sampling time point (t2). Patient 10 remained positive

reference strains COL (GenBank accession no. NC_002951) and N315 (GenBank accession no. BA000018.3) were included in the SeqSphere+

dendrogram. Isolates (BU_G0201_t8 and BU_G0202_t2 belonging to CT 545 and BU_G1074_t4 and BU_G1001_t8 belonging to CT 547) originating frompatients 2 and 10 at two different time points are included in both transmission events. NA means ‘Not Applicable’.

doi:10.1371/journal.pntd.0004049.g001

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 6 / 12

Tab

le2.

Gen

otypic

andphen

otypic

charac

teristicsof2

1S.a

ureu

sisolatesfrom

BUpatients.

Patient

No.

Sam

ple

IDStartof

trea

tmen

tEndof

wound

care

Wee

kof

samplin

gMLVF

CT

ST

spa-

type

SCCmec

CXT

OXA

CIP

TET

CHL

TRI

STM

RIF

2BU_G

0201

_t8

1-12

-201

2Not

yet

8H

545

88t786

IVa

+mec

AtetM

,tetL

fexB

7BU_G

0701

_t5

26-12-20

1225

-6-201

35

H54

588

t786

IVa

+mec

AtetM

,tetL

fexB

2BU_G

0202

_t2

1-12

-201

2Not

yet

2H

545

88t786

IVa

+mec

AtetM

,tetL

fexB

19BU_G

1905

_t3

5-1-20

1322

-7-201

33

H54

588

t786

IVa

+mec

AtetM

,tetL

fexB

13BU_W

13_t1

12-12-20

1226

-4-201

31

J71

488

t186

IVa

+mec

AtetM

,tetL

fexB

22BU_W

22_t4

12-12-20

12Not

yet

4I

556

5t272

4tetK

drfG

7BU_W

7A_t11

26-12-20

1225

-6-201

311

K54

65

t002

none

++

gyrA*

17BU_N

17Y_t2

28-12-20

12N/A

2B

552

15t346

drfG

3BU_N

3_t2

6-12

-201

2N/A

2A

554

15t084

tetM

6BU_W

6_t1

12-12-20

1226

-4-201

31

G71

51

t783

5tetK

fexB

12BU_G

1201

_t13

19-12-20

1225

-6-201

313

L54

912

1t314

tetK

catA

drfG

12BU_G

1201

_t8

19-12-20

1225

-6-201

38

L54

912

1t314

tetK

catA

drfG

26BU_G

2601

A_t9

6-3-20

1321

-6-201

39

L55

012

1t314

tetK

drfG

rpoB

*

22BU_N

22_t6

12-12-20

12N/A

6M

553

3019

t939

catA

12BU_W

12_t13

19-12-20

1225

-6-201

313

C55

550

8t128

36

3BU_G

0301

_t8

6-12

-201

2Not

yet

8F

547

152

t355

tetK

catA

str

10BU_G

1074

_t4

12-12-20

1226

-4-201

34

F54

715

2t355

tetK

catA

str

11BU_G

1101

_t2

19-12-20

125-3-20

132

F54

715

2t355

tetK

10BU_G

1001

_t8

12-12-20

1226

-4-201

38

F54

715

2t335

tetK

catA

str

17BU_N

17W_t2

28-12-20

12N/A

2E

551

152

t113

75

7BU_G

0706

B_t8

26-12-20

1225

-6-201

38

D54

815

2t109

6tetK

CXT—ce

foxitin

,OXA—ox

acillin,C

IP—ciprofl

oxac

illin,T

ET—tetrac

ycline,

CHL–

chloramph

enicol,T

RI–trim

etho

prim

,STM–streptom

ycin

andRIF—rifam

picin.

The

‘*’symbo

lind

icates

amutationin

aco

rege

nomege

nethat

isinvo

lved

inan

tibiotic

resistan

ce.A

llge

nesareinvo

lved

insp

ecificresistan

ceph

enotyp

es.

doi:10.1371/journal.pntd.0004049.t002

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 7 / 12

for S. aureus with the CT 547 until sampling time point t8, and patient 11 until time point t5(Table 1). A third patient (patient 3) was found to be positive for S. aureus with the CT 547 atsampling time point t8 (Table 1). Patients 5, 6, 7, 18 and 24 became positive for S. aureus withthis genotype at later time points than patients 10 and 11. The patients 5, 6, 7, 18 and 24 paid atleast one visit at the health center for wound care that overlapped with visits by three otherpatients, which were found to be positive for the S. aureus genotype with the CT 547 (Table 1).

It is noteworthy to mention that in each of the transmission events, the gene allele variationbetween isolates was not higher than one. This implies that the isolates were nearly identicalwith respect to their core genome.

Antibiotic resistance genesThe assembled genomes of the 21 S. aureus isolates were used in blast comparisons to detectthe presence of AMR genes, and the results are shown in Table 2. Among the investigated

Fig 2. Minimum spanning tree showing the allelic difference between 21 S. aureus isolated from BU patients. The tree was generated using theSeqSphere+ software. The node size in the tree is proportional to the frequency of genotype occurrence. The allelic difference between S. aureus isolates isindicated as numbers between each node. Isolates with blue and red circles belonging to CTs 545 (MLVF cluster H) and 547 (MLVF cluster F) confirmtransmission events. Isolates with green and yellow circles belonging to CTs 549 and 550 were initially grouped in MLVF cluster L and do not confirmtransmission events.

doi:10.1371/journal.pntd.0004049.g002

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 8 / 12

isolates none was found to carry genes involved in resistance to erythromycin, fusidic acid,kanamycin, mupirocin, or vancomycin. Antibiotic resistance of the sequenced isolates was pre-viously most often found against penicillin, chloramphenicol, tetracycline and trimethoprim[3]. Consistent with their penicillin resistance, all sequenced isolates carried various types ofblaZ operons, which were located on chromosomally integrated transposons or plasmids. Spe-cifically, the blaZ gene was found in 16 isolates that belonged to ST1, ST5, ST15, ST88, ST152and ST3019, while the blaZ-B variant was found in five isolates representing ST5, ST508 andST121. Fourteen sequenced S. aureus isolates were chloramphenicol resistant of which six(ST121, ST3019 and ST152) carried various plasmids with a catA gene. Six other isolates (ST88and ST1) carried a novel allele of fexB that was not previously reported in S. aureus. In the caseof one isolate, the phenotypic resistance for chloramphenicol could not be confirmed at thegenomic level, which was potentially due to the loss of the resistance gene. Resistance to rifam-picin was identified in one isolate belonging to ST121 where the rpoB gene was found to encodean amino acid substitution that changed Asp471 into Gly. Resistance to tetracycline was identi-fied in 16 isolates, which was confirmed by the identification of resistance genes, such as tetK,tetL and tetM. The tetK gene was located on plasmid pT181, which was found in 10 isolatesrepresenting different STs. Five isolates of ST88 contained the tetL and tetM genes located onidentical mobile genetic elements integrated into their genomes, while one isolate of ST15 con-tained a transposon with tetM. The presence of a plasmid or transposon carrying the drfG generesponsible for trimethoprim resistance was detected in five isolates that belonged to ST5,ST15 and ST121. Resistance to streptomycin was limited to three isolates of ST152 where thestr gene was present. Of the six methicillin resistant isolates, five belonging to ST88 containedthemecA gene, whereas one ST5 isolate contained neithermecA normecC. The latter isolatewas termed borderline oxacillin resistant S. aureus (BORSA). Intriguingly, the BORSA isolatecontained no mutations in the genes for the penicillin-binding proteins PBP1, PBP2, and PBP3or the YjbH protein, which were previously proposed to be involved in BORSA phenotypes[24]. However, sequence comparisons revealed that the PBP2 protein of the BORSA isolatecontains a Tyr residue at position 197, while the PBP2 protein of S. aureusN315 contains a Cysresidue at this position. Furthermore, the BORSA isolate showed resistance to fluoroquino-lones, which may be due to a specific mutation in the gyrase A gene (Ser84Leu).

DiscussionIn the present study, we have investigated S. aureus transmission events in BU patients duringwound care by implementing a WGS-based gene-by-gene typing approach using SeqSphere+.The SeqSphere+ scheme grouped the 21 sequenced S. aureus isolates into eight different STs.Sequenced S. aureus that belonged to ST88 isolates shared identical characteristics (spa-typest186/t786, SCCmec type IVa, PVL-negative) with isolates collected from out-patients in Egyptand Angola, indicating a larger geographic distribution on the African continent [25,26]. Thenew ST3019 (spa-type t939) identified in this study belongs to the same clonal complex(CC45) as ST45 and ST508. Compared to ST45 a single locus variation was observed at theyqiL locus for ST3019 and in the aroE locus for ST508.

Using SeqSphere+, we identified two major clusters of S. aureus isolates from different BUpatients, which may reflect transmission events that occurred during overlapping visits to thePakro Healthcare Center where these patients received wound care. None of these patients car-ried S. aureus with the CTs 545 or 547 on their first visit to the Pakro Healthcare Center,strongly suggesting that they acquired the respective S. aureus types upon wound care. Interest-ingly, the majority of S. aureus isolates from BU patients belong to lineages characterized byspa-types t786 and t355 that have been already reported in health care settings in Ghana [11].

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 9 / 12

This suggests the nosocomial acquisition of these S. aureus types by patient-to-patient trans-mission between BU patients and healthcare workers that may have occurred due to inade-quate hygiene. Indeed, it has been reported in a recent study that 8 of 11 MRSA transmissionevents among patients in intensive care settings were potentially due to poor hand hygiene[16]. This could probably be avoided by wearing gloves and protective gowns, and strict imple-mentation of hand hygiene [27–29]. Basic preventive measures, such as adherence to aseptictechniques may further reduce the risk of infection thereby improving wound care of patients,provided that gloves, gowns, adequate dressing materials, running water and hand rub alcoholare made available. With a steady supply and stock of equipment and disposables, routinescreening of patients and healthcare workers for S. aureusmay be less critical.

Genotypic data of the isolates sequenced confirmed the results of the antimicrobial resistanceprofiles described previously [3]. Interestingly, the chloramphenicol resistance of some isolateswas conveyed by the fexB gene (Table 2), which was thus far not encountered in S. aureus. Onthe other hand, fexBwas previously reported in Enterococcus faecium EFM-1 and Enterococcushirae EH-1 isolates from pigs [30]. As Enterococci were previously identified in the wounds ofBU patients, it is conceivable that the MRSA isolates acquired the fexB gene by horizontal genetransfer from such Enterococci [3]. Furthermore, a BORSA phenotype was identified in an isolatebelonging to ST5. Such a BORSA phenotype was previously reported for S. aureus isolates withST1, ST8 and ST15 that were implicated in wound infections in Scotland [24]. The presence ofspecific mutations in the genes coding for four proteins, namely PBP1, PBP2, PBP3 and YjbH,was proposed to be involved in the BORSA phenotype. However, in the genome sequence of thepresently investigated BORSA isolate from a BU patient, none of these mutations was found.After genomic comparison of the BORSA isolate with the N315 reference genome, the only dif-ference was observed for PBP2, where at position 197 a cysteine residue was replaced by a tyro-sine residue. However, this PBP2 amino acid substitution is encoded by the majority of S. aureusgenomes available in the NCBI database and, therefore, it may not explain the BORSA phenotypeobserved. Further comparative genome analyses revealed about 300 additional non-synonymousSNPs, which could contribute to the observed BORSA phenotype.

In summary, WGS of S. aureus isolates from BU patients and the subsequent analysis ofsequencing data using the SeqSphere+ scheme revealed likely patient-to-patient transmissionevents in a healthcare setting in Ghana. This indicates a need for the implementation of improvedhygiene protocols in healthcare settings where BU patients receive wound care. Apart from thedetection of transmission events, WGS has the advantage that it also provides information onantimicrobial resistance. Related to the antimicrobial resistance pheno- and genotypes identifiedin S. aureus isolates from BU patients, it is important to bear in mind that antimicrobial pressurehas the potential to aggravate resistance, with an inherent risk for transmission of resistant organ-isms. Therefore, even in low-resource settings, antimicrobial stewardship programs are likely tohave added value, with more restrictive antimicrobial use than currently practiced [2].

Supporting InformationS1 Table. Information on reference genomes from which the antibiotic resistance geneswere queried.(DOCX)

AcknowledgmentsThe authors express their gratitude to the Buruli ulcer health team from the Pakro Health Cen-ter (Ghana) for their help in sample collection and treatment of the patients.

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 10 / 12

Author ContributionsConceived and designed the experiments: NAA AA TSvdW JMvD JWR YS. Performed theexperiments: NAAMAC AJS CST IP. Analyzed the data: NAAMAC JWR. Contributedreagents/materials/analysis tools: MAC JMvD AA AWF TSvdW JWR YS. Wrote the paper:NAAMAC AJS JMvD TSvdW JWR YS.

References1. Jassens PG; Pattyn SR; Meyers WM; Portaels F (2005) Buruli ulcer: An historical overview with updat-

ing to 2005. Bull Séanc Acad R Sci O-m 51: 165–199.

2. Barogui YT, Klis S, Bankole H, Sopoh GE, Mamo S, et al. (2013) Towards Rational Use of Antibioticsfor Suspected Secondary Infections in Buruli Ulcer Patients. PLoS Negl Trop Dis 7.

3. Amissah NA, Glasner C, Ablordey A, Tetteh CS, Kotey NK, et al. (2015) Genetic Diversity of Staphylo-coccus aureus in Buruli Ulcer. PLoS Negl Trop Dis 9: e0003421. doi: 10.1371/journal.pntd.0003421PMID: 25658641

4. Van Belkum A, Melles D, Nouwen J, van LeewenW, vanWamel W, et al. (2009) Co-evolutionaryaspects of human colonisation and infection by Staphylococcus aureus. Infect Genet Evol 9: 32–47.http://www.sciencedirect.com/science/article/pii/S1567134808001779. Accessed 30 April 2014. doi:10.1016/j.meegid.2008.09.012 PMID: 19000784

5. Lowy FD (1998) Staphylococcus aureus infections. N Engl J Med 339: 520–532. PMID: 9709046

6. Kluytmans J, van Belkum A, Verbrugh H (1997) Nasal carriage of Staphylococcus aureus: epidemiol-ogy, underlying mechanisms, and associated risks. Clin Microbiol Rev 10: 505–520. PMID: 9227864

7. Aiken AM, Mutuku IM, Sabat AJ, Akkerboom V, Mwangi J, et al. (2014) Carriage of Staphylococcusaureus in Thika Level 5 Hospital, Kenya: a cross-sectional study. Antimicrob Resist Infect Control 3:22. http://www.ncbi.nlm.nih.gov/pubmed/25057351. doi: 10.1186/2047-2994-3-22 PMID: 25057351

8. Tigist Alebachew, Gizachew Y, Derabe A, Sisay Z (2012) Staphylococcus aureus burn wound infectionamong patients attending Yekatit 12 Hospital burn unit, Addis Ababa, Ethiopia. Ethiop J Health Sci 22:209–213. PMID: 23209356

9. Van der Kooi-Pol MM, Veenstra-Kyuchukova YK, Duipmans JC, Pluister GN, Schouls LM, de NeelingAJ, Grundmann H, JonkmanMF van DJ (2012) High genetic diversity of Staphylococcus aureus strainscolonizing patients with epidermolysis bullosa. Exp Dermatol 21: 463–466. doi: 10.1111/j.1600-0625.2012.01502.x PMID: 22621190

10. Van der Kooi-Pol MM, Sadaghian SadabadM, Duipmans JC, Sabat AJ, Stobernack T, et al. (2013)Topography of Distinct Staphylococcus aureus Types in Chronic Wounds of Patients with Epidermoly-sis Bullosa. PLoS One 8.

11. Egyir B, Guardabassi L, SørumM, Nielsen SS, Kolekang A, et al. (2014) Molecular Epidemiology andAntimicrobial Susceptibility of Clinical Staphylococcus aureus from Healthcare Institutions in Ghana.PLoS One 9: e89716. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3934920&tool =pmcentrez&rendertype = abstract. doi: 10.1371/journal.pone.0089716 PMID: 24586981

12. McBryde ES, Bradley LC, Whitby M, McElwain DLS (2004) An investigation of contact transmission ofmethicillin-resistant Staphylococcus aureus. J Hosp Infect 58: 104–108. PMID: 15474180

13. Otter JA, Yezli S, French GL (2011) The role played by contaminated surfaces in the transmission ofnosocomial pathogens. Infect Control Hosp Epidemiol 32: 687–699. doi: 10.1086/660363 PMID:21666400

14. Dancer SJ, White LF, Lamb J, Girvan EK, Robertson C (2009) Measuring the effect of enhanced clean-ing in a UK hospital: a prospective cross-over study. BMCMed 7: 28. doi: 10.1186/1741-7015-7-28PMID: 19505316

15. Simor AE, Daneman N (2009) Staphylococcus aureus Decolonization as a Prevention Strategy. InfectDis Clin North Am 23: 133–151. doi: 10.1016/j.idc.2008.10.006 PMID: 19135919

16. Moore G, Cookson B, Gordon NC, Jackson R, Kearns A, Singleton J, Smyth DWA (2015) Whole-genome sequencing in hierarchy with pulsed-field gel electrophoresis: the utility of this approach toestablish possible sources of MRSA cross-transmission. J Hosp Infect S0195-6701: 00032–00038.

17. Tong SY, Holden MT, Nickerson EK, Cooper BS, Köser CU, Cori A, Jombart T, Cauchemez S, FraserC, Wuthiekanun V, Thaipadungpanit J, HongsuwanM, Day NP, Limmathurotsakul D, Parkhill J PS(2015) Genome sequencing defines phylogeny and spread of methicillin-resistant Staphylococcusaureus in a high transmission setting. Genome Res 25: 111–118. doi: 10.1101/gr.174730.114 PMID:25491771

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 11 / 12

18. Long SW, Beres SB, Olsen RJ MJ (2014) Absence of patient-to-patient intrahospital transmission ofStaphylococcus aureus as determined by whole-genome sequencing. MBio 5: e01692–14. doi: 10.1128/mBio.01692-14 PMID: 25293757

19. Sabat AJ, Chlebowicz MA, Grundmann H, Arends JP, Kampinga G, et al. (2012) Microfluidic-Chip-Based Multiple-Locus Variable-Number Tandem-Repeat Fingerprinting with New Primer Sets for Meth-icillin-Resistant Staphylococcus aureus. J Clin Microbiol 50: 2255–2262. doi: 10.1128/JCM.00056-12PMID: 22573591

20. Sabat a. J, Budimir a., Nashev D, Sá-Leão R, van Dijl J, et al. (2013) Overview of molecular typingmethods for outbreak detection and epidemiological surveillance. Euro Surveill 18: 20380. PMID:23369389

21. Karynski M, Sabat AJ, Empel J, Hryniewicz W (2008) Molecular surveillance of methicillin-resistantStaphylococcus aureus by multiple-locus variable number tandem repeat fingerprinting (formerly multi-ple-locus variable number tandem repeat analysis) and spa typing in a hierarchic approach. DiagnMicrobiol Infect Dis 62: 255–262. doi: 10.1016/j.diagmicrobio.2008.06.019 PMID: 18708278

22. Leopold SR, Goering R V., Witten A, Harmsen D, Mellmann A (2014) Bacterial whole-genomesequencing revisited: Portable, scalable, and standardized analysis for typing and detection of viru-lence and antibiotic resistance genes. J Clin Microbiol 52: 2365–2370. doi: 10.1128/JCM.00262-14PMID: 24759713

23. Carver TJ, Rutherford KM, Berriman M, RajandreamMA, Barrell BG, et al. (2005) ACT: The Artemiscomparison tool. Bioinformatics 21: 3422–3423. PMID: 15976072

24. Ba X, Harrison EM, Edwards GF, Holden MTG, Larsen AR, et al. (2014) Novel mutations in penicillin-binding protein genes in clinical Staphylococcus aureus isolates that are methicillin resistant on sus-ceptibility testing, but lack the mec gene. J Antimicrob Chemother 69: 594–597. doi: 10.1093/jac/dkt418 PMID: 24216768

25. Abou Shady HM, Bakr AEA, Hashad ME, Alzohairy MA (2015) Staphylococcus aureus nasal carriageamong outpatients attending primary health care centers: a comparative study of two cities in SaudiArabia and Egypt. b raz j inf e c t di s 19: 68–76.

26. Conceição T, Coelho C, Santos-Silva I, de Lencastre H A-SM (2014) Epidemiology of methicillin-resis-tant and-susceptible Staphylococcus aureus in Luanda, Angola: first description of the spread of theMRSA ST5-IVa clone in the African continent. Microb Drug Resist 20: 441–449. doi: 10.1089/mdr.2014.0007 PMID: 24694289

27. Derde LPG, Cooper BS, Goossens H, Malhotra-Kumar S, Willems RJL, et al. (2014) Interventions toreduce colonisation and transmission of antimicrobial-resistant bacteria in intensive care units: An inter-rupted time series study and cluster randomised trial. Lancet Infect Dis 14: 31–39. doi: 10.1016/S1473-3099(13)70295-0 PMID: 24161233

28. Harris AD, Pineles L, Belton B, Johnson JK, Shardell M, et al. (2013) Universal glove and gown useand acquisition of antibiotic-resistant bacteria in the ICU: a randomized trial. JAMA 310: 1571–1580.http://www.ncbi.nlm.nih.gov/pubmed/24097234. PMID: 24097234

29. Sopirala MM, Yahle-Dunbar L, Smyer J, Wellington L, Dickman J, et al. (2014) Infection control linknurse program: An interdisciplinary approach in targeting health care-acquired infection. Am J InfectControl 42: 353–359. doi: 10.1016/j.ajic.2013.10.007 PMID: 24548456

30. Liu H, Wang Y, Wu C, Schwarz S, Shen Z, et al. (2012) A novel phenicol exporter gene, fexB, found inenterococci of animal origin. J Antimicrob Chemother 67: 322–325. doi: 10.1093/jac/dkr481 PMID:22096043

Transmission Events of Staphylococcus aureus in Buruli Ulcer Patients

PLOS Neglected Tropical Diseases | DOI:10.1371/journal.pntd.0004049 September 11, 2015 12 / 12